Endometrial cancer is the most common gynecological cancer, with more than 69,000 cases diagnosed in the United States in 2025 and an increase of up to 3% each year. Diagnosis often requires a painful and invasive biopsy, which carries the risk of false negatives. An interdisciplinary research team at Washington University in St. Louis and Siteman Cancer Center based at Barnes-Jewish Hospital and WashU Medicine is focusing on fast, safe, and non-invasive imaging methods combined with machine learning for the accurate detection and diagnosis of precancerous lesions and early cancers.
A team led by Quinn Zhu, the Edwin H. Murthy Professor of Engineering at Washington University’s McKelvey School of Engineering in St. Louis, conducted the initial investigation using optical coherence tomography (OCT), which detects differences in the way tissue reflects light and obtains high-resolution 3D images up to 1 to 2 millimeters deep. Using a custom catheter probe developed in Zhu’s lab, the team was able to image the entire endometrial cavity and create an optical biopsy within three seconds. This is the first catheter-based 3D OCT imaging study that integrates optical, structural and radiological features for endometrial assessment. The research results are npj imagineg June 3, 2026.
To obtain images from patients’ tissues, the team collaborated with WashU Medicine physicians led by Lindsey Kloki, MD, associate professor of obstetrics and gynecology, and Ian Hageman, MD, professor of pathology and immunology and obstetrics and gynecology. Together with Zhu, they are research members at the Siteman Cancer Center, where Kuroki also treats patients. In 2025, the research team acquired OCT images of 57 uteri after hysterectomy. Thirty-four of these cases included high-risk precancerous lesions or early-stage cancer.
3D OCT images showed tissue microstructure and optical properties in detail, revealing distinct differences between normal endometrium, benign endometrium, high-risk precancerous lesions, and different stages of endometrial cancer.
First authors Sanskar Thakur, a PhD student in Zhu’s lab, and Yixiao Lin, who earned his PhD in biomedical engineering at WashU in 2025, used the 26 extracted image features to develop an image feature extraction pipeline and machine learning model to classify the results into two groups: normal and benign, and precancerous and cancerous. Their model achieved a search sensitivity of 94% and specificity of 87%.
The current estimated false-negative rate for endometrial biopsies is approximately 10% (sensitivity approximately 90%), which is primarily due to sampling limitations and variability in interpretation. By combining our 3D OCT imaging system with machine learning, the entire endometrial cavity can be imaged in 2-3 seconds, potentially achieving higher sensitivity than random biopsy sampling. ” said Zhu.
Quing Zhu, Edwin H. Marty Professor of Engineering, McKelvey School of Engineering, Washington University in St. Louis
“Currently, there is no reliable screening for endometrial cancer,” said co-author David Mutch, the Ira C. and Judith Gall Professor and vice chair of obstetrics and gynecology at WashU Medicine, Siteman study member, and principal investigator of the National Cancer Institute-funded Route 66 Special Research Program on Endometrial Cancer (SPORE) grant. “This technology developed by Dr. Zhu and her colleagues should allow us to better screen for this cancer, or at least detect it much earlier in its development,” Match added. “This is truly novel and cutting-edge technology.”
Zhu said the team will now evaluate the catheter in live patients to demonstrate the translational potential of AI-assisted OCT technology.
sauce:
Washington University in St. Louis
Reference magazines:
Thakur, S. others. (2026). Optical coherence tomography allows optical biopsy of endometrial tissue for early cancer detection. npj imaging. DOI: 10.1038/s44303-026-00160-z. https://www.nature.com/articles/s44303-026-00160-z
